Development of new drug therapies for CNS disorders has significantly lagged behind other indication areas, with some estimates suggesting only a 1% success rate for new chemical entities in the United States. Factors contributing to this problem include: lack of disease-relevant functional screens, lack of clinically predictive animal models and absence of reliable and specific biomarkers of disease state. These challenges have particularly impacted the discovery of cognitive therapies for schizophrenia. The cognitive impairments in schizophrenia comprise a core and debilitating component of the illness for which there are currently no effective therapies. Development of animal models that can reliably mirror human cognition is challenging, since the cognitive capacities and behavioral repertoires of animals and humans are fundamentally distinct. One approach to bridging this gap is to define specific aspects of neural activity that are altered in the human disease state and in cognate animal models, and to include such measures in testing candidate drug efficacy. The power of this approach is that it takes advantage of phylogenetic conservation at the level of neural networks to translate directly between species. For example, in accord with previous reports noting a severe dysfunction in the ability of cortical networks to mount coherent gamma oscillations in schizophrenia patients, we have observed alterations in gamma oscillations in the calcineurin knockout (CNKO) and phencyclidine- mediated mouse models of schizophrenia. We observed significant modulation of gamma oscillations in the prefrontal cortex of mice during mnemonic tasks associated with novelty recognition/detection that is impaired in these mouse disease models. In an initial study aimed at identifying human equivalencies of this novelty- related neural activity, we have observed similar changes in gamma oscillations within frontal cortex during a novelty oddball task in a small population of healthy volunteers. These data provide support for the potential utility of neurophysiological endophenotypes within defined neural circuits as objective measures of cognitive disease states that can be translated from animal models to human patients. We propose to identify and validate cortical neurophysiological endophenotypes in both humans and mice that are associated with analogous cognitive behavioral tasks and similarly affected by disease state. A coordinated and reciprocal approach will be applied since human studies have the advantage of direct disease relevance, while rodent systems can provide higher resolution electrophysiological measures and can be manipulated via genetic and pharmacological means to test specific disease hypotheses. We propose that identified electrophysiological endophenotypes of the disease condition that are conserved between rodents and man will provide objective biomarkers for schizophrenia disease state and for assessing drug candidate efficacy during clinical trials, as well as diagnostic tools for personalization of optimal treatment regimes for patients. The research outlined in this proposal thus stands to benefit the large population of patients with schizophrenia and related illnesses.